“On the Persistence of Persistent Identifiers of the Scholarly Web” is a new paper from Los Alamos, finding that many DOIs in a 10,000 random sample are unreachable…

“consistently across request methods, more than half of our DOIs fail to successfully resolve to a target resource”

Despite the misleading “2004” tag on the page identifier tag, the paper was actually presented in March 2020 at the CNI Spring 2020 Project Briefings.